Model specifications
- Context
- 131,072 tokens
- Maximum output
- 16,384 tokens
- Architecture
- text+image->text
- Tokenizer
- Other
- Knowledge cutoff
- Not provided
- Moderated
- No
Input → Output
No · xhigh · high · medium · low
Supported API parameters
frequency_penaltyinclude_reasoninglogit_biaslogprobsmax_tokensmin_ppresence_penaltyreasoningreasoning_effortrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p5 providers
Phala
unknown- Context
- 131K
- Maximum output
- 118K
- Input
- $0.300
- Output
- $1.10
- Cache read
- $0.040
- Cache write
- Not provided
Parasail
bf16- Context
- 131K
- Maximum output
- 118K
- Input
- $0.300
- Output
- $1.10
- Cache read
- $0.040
- Cache write
- Not provided
DeepInfra
bf16- Context
- 131K
- Maximum output
- 16K
- Input
- $0.300
- Output
- $1.20
- Cache read
- $0.040
- Cache write
- Not provided
Together
unknown- Context
- 131K
- Maximum output
- 118K
- Input
- $0.350
- Output
- $1.50
- Cache read
- $0.040
- Cache write
- Not provided
Fireworks
unknown- Context
- 131K
- Maximum output
- 118K
- Input
- $0.350
- Output
- $1.50
- Cache read
- $0.040
- Cache write
- Not provided